Adult male Sprague Dawley rats (Charles Rivers Laboratories) weighing 300–350 gm were housed in pairs in plastic cages and maintained on a 12 h light/dark cycle. Food and water were provided ad libitum throughout the experiment. Animal use procedures were in strict accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and were approved by the University of Texas at Dallas Animal Care and Use Committee.

Electrical stimulation experiments were performed as previously described (Ploski et al., 2008, 2010). For LTP stimulation experiments, Sprague Dawely rats (300–350 g) were anesthetized with urethane (2 i.p. injections at 10 min intervals; total of 1.6 mg/kg) and placed in a stereotaxic frame. The skull was exposed and the rats were implanted with a concentric bipolar stimulating electrode (Kopf Instruments model #NEX-100) into the angular bundle of the perforant path (-7.8 AP, 4 ML, -3.4 DV). One-half hour following implantation of the stimulating electrode, rats were given LTP-inducing HFS, which consisted of six trains of pulses (400 Hz, 20 ms/pulse), delivered at a 10 s interval and repeated six times at an interval of 2 min with a stimulation intensity of 500 μAmps, 100 μs. This protocol has been widely used in the perforant-dentate pathway to produce reliable and robust potentiation of perforant path synapses (Davis et al., 2000; Messaoudi et al., 2007; Ploski et al., 2010). In all stimulation experiments current was applied such that it traveled from the tip to the tube of the bipolar stimulation electrode. The rats were sacrificed at either 2 or 4 h following HFS and the brain was dissected and immediately frozen on powdered dry ice and stored at -80°C until further processing.

At the appropriate time point, the brains were rapidly dissected and promptly frozen with powdered dry ice and stored at -80°C until further processing. Twenty-micron coronal sections containing the anterior dorsal hippocampus were obtained and rapidly frozen immediately after being mounted on Fisher Superfrost slides. Sections were fixed in 4% PFA in PB Buffer (NaH₂PO₄ monohydrate 125 mM, NaOH 96 mM) for 10 min. Sections were then washed 3x for 10 min in PBS-A (NaCl 150 mM, NaOH 96 mM, NaH₂PO₄ 125 mM), followed by a 1 h incubation in blocking solution 1% bovine serum albumin (1% BSA, Sigma, Cat #A-3059; 0.1% Triton X-100, AmericanBio, #AB02025), slices were incubated overnight at room temperature in anti-Arc/Arg3.1 antibody (1:500; Santa Cruz, #17839) in blocking solution, at room temperature in a humid chamber. On the following day sections were washed 3x with PBS-B (NaCl 150 mM, NaOH 10 mM, NaH₂PO₄ 12.5 mM) followed by a 2 h incubation in Texas Red secondary antibody (1:1000, Life Technologies, #T6390). Sections were then washed 3x with PBS-B and mounted (VectaShield, H-1000). IHC images were captured at 100× magnification using an Olympus BX51 upright fluorescence microscope with an Olympus DP71 Digital Camera and DP manager software. Only samples exhibiting robust and uniform expression of Arc in the granule and molecular layers for the stimulated side of the brain were selected for further study. We attempted to deliver pp-HFS to 43 animals of which we found that 23 exhibited robust Arc IHC signal within the dentate gyrus and these were the animals used for further study.

Dissection and purification procedures were performed as previously described in Partin et al. (2013). Ten-micron sections containing the dorsal hippocampus were taken and placed onto MMI Laser Microdissection (LMD) slides (MMI, #50102) and stored at -80°C until further processing. Immediately prior to microdissection, the slides were subjected to ethanol dehydration (75% 30 s, 95% 30 s, 100% 30 s, xylene 30 s, xylene 5 min) using Histogen LCM Frozen Section Staining Kit (Arcturus, Mountain View, CA, United States). Microdissection was performed via laser microdissection on a SmartCut Laser Microdissection System, which was mounted on an Olympus CKX41 inverted microscope. For samples that were used in microarray analysis, 120 sections were dissected per sample (control, n = 3; 2 h, n = 3; 4 h, n = 3) from the anterior-medial dentate gyrus, in which the distal 2/3rds of the molecular layer was collected. For samples utilized in qRT-PCR analysis, another set of animals were stimulated, dissected and subjected to qRT-PCR analysis. These animals had the distal 1/3rd of the molecular layer collected, as well as the granule layer across twenty-one 10 micron sections per sample (control, n = 5; 2 h, n = 4; 4 h, n = 4). Control samples consisted of the contralateral, non-stimulated molecular and granule layers, respectively. Samples were collected in 25 μL of cell lysis buffer and purified using RNAqueos-Micro Kit (Ambion, #AM1931). Samples were re-purified via precipitation using Pellet Paint NF (Novagen) and resuspended in a 10 μL volume. All steps were carried out according to manufacturers’ instructions. Purified RNA was converted to cDNA (SuperScript ll; Invitrogen, #18064014) for qRT-PCR analysis or amplified for microarray analysis.

RNA in a 10 μL volume was amplified using MessageAmp II aRNA Amplification Kit (Ambion) according to the manufacturer’s instructions. The amplified RNA from this first round of RNA amplification was subjected to a second round of RNA amplification using MessageAmp II-Biotin Enhanced aRNA Amplification Kit and was performed according to the manufacturer’s instructions to yield Biotin labeled aRNA suitable for Affymetrix microarray analysis. RNA from second round amplification was also converted to cDNA for qRT-PCR analysis.

Was essentially performed as previously described (Ploski et al., 2006, 2010; Partin et al., 2013). A total of 9 microarray hybridizations were performed at the UTSW microarray facility using Affymetrix single channel Rat Genome 230 2.0 Arrays, n = 3 per group (3x unstimulated, 3 × 2 h post-stimulation and 3 × 4 h post-stimulation). Gene lists were created based on the relatively stringent criteria that the gene must exhibit an average fold difference of threefold or greater in pair wise comparisons between the unstimulated and stimulated samples, with a t-test p-value of p < 0.05. Importantly, the Microarray Quality Control (MAQC) Consortium has reported that this approach can be successful in identifying reproducible gene lists (Shi et al., 2006). Supplemental data including p-values, gene bank accession numbers and full gene names, etc. are included in Supplementary File S1.

Quantitative RT-PCR was performed using the ΔΔC_t method as we have described previously (Ploski et al., 2006, 2008, 2010; de Solis et al., 2015) using a CFX96 Real-Time System thermocycler (BioRad, #1845096) and QuantiTect SYBR Green PCR Kit (Qiagen, #204143) with custom designed primers at a concentration of 300 nM. For qRT-PCR experiments, all samples were performed in duplicate and relative gene concentrations were normalized against GAPDH levels. qRT-PCR was performed with the following conditions [(95°C for 15 min) ((94°C for 30 s, 55°C for 30 s, 72°C for 30 s) × (35 cycles))] in a standard 20 μL reaction. PCR primers utilized for qRT-PCR were examined for their efficiency of PCR amplification and were found to be ∼99% efficient. Primer sequences can be found in Supplementary File S2.

Standard PCR was performed using Titanium Taq (Clontech Laboratories, #639208) in a 20 μL volume from rat cDNA from the molecular layer using the following conditions [(95°C for 60 s) ((95°C for 30 s, 68°C for 60 s, 72°C for 30 s) × (35 cycles))].

PCR reactions were performed with the following primer sets: (miR-212 FP TAACAGTCTCCAGTCACGGCCACCGACGCC; RM2 RP GGTCTCACTGTAGTTCTGGCTAGCCTTGAACTCACAGAAACCC) (miR-132 FP CAGGGCAACCGTGGCTTTCGATTGTTACTGTGGGAACCGG; RM2 RP). PCR products of 1.5 kb and ∼0.6 kb were obtained for the PCR amplification using the miR-132 FP and RM2 RP primers. No PCR products were obtained using the miR-212 FP and RM2 RP. The PCR products were cloned into the pCR4-Topo vector via a TA cloning kit (Invitrogen, #450030) and sequenced (Retrogen, Inc.).

The probe template for Arc was a generous gift from Oswald Steward and it was designed to target the entire Arc coding region (Guzowski et al., 1999; Farris et al., 2014). The probe template for Homer1a was designed to target the Homer1a 3′ untranslated region and was a generous gift from John Guzowski (Guzowski et al., 1999; Vazdarjanova et al., 2002). The DNA templates to produce fluorescent in situ hybridization (FISH) RNA probes for Egr3, Egr4, Sox11, Ptgs2, pri-miR132 were PCR amplified from rat cDNA and these PCR products were cloned into the pCR4-TOPO vector using a TA cloning kit (Invitrogen, #450030). FISH Probes were prepared using MAXIscript T7/T3 Transcription Kit (Ambion, #AM1324) and were labeled with UTP-Digoxigenin (Roche, #11209256910). These probes were purified using mini Quick Spin RNA Columns (Roche, #11814427001). RNA probes that were labeled with radioactivity was performed as previously described (Ploski et al., 2010; Partin et al., 2013). DNA templates for Arc and Nurr1 were PCR amplified and used to produce a radiolabeled probe using a T7-based in vitro transcription kit (Megashortscript; Ambion) using [³⁵S]CTP (1.5 μCi) (PerkinElmer). Removal of unincorporated nucleotides after the in vitro transcription reaction was performed using sepharose spin columns (Roche). PCR primers used to generate DNA templates for RNA probe production are listed in Supplementary File S2.

In situ hybridization was performed as previously described (Ploski et al., 2010; Partin et al., 2013). FISH was performed as previously described (Farris et al., 2014). For FISH, Prior to hybridization, sections were fixed in 4% PFA in PBS (pH 7.4) for 5 min, rinsed in 2x SSC (AmericanBio, #AB13156) for 2 min, incubated in 0.50% acetic anhydride (Sigma, #320102) in 1.5% triethanolamine (Sigma, #90276) for 10 min and then treated with 1:1 acetone (Sigma, #270725): methanol (Fisher, #A412) for 5 min. Prehybridization was performed at 56°C for 1 h in hybridization buffer [(50% formamide (AmericanBio, #AB00600), 5x SSC, 1.25x denhardt’s solution (AmericanBio, #AB03075), 250 μg/mL, E. coli tRNA (Sigma, #R1753), 500 μg/mL salmon sperm (Sigma, #D7656) and 5% dextran sulfate (AmericanBio, # AB00426)]. Sections were hybridized overnight (12–14 h) in hybridization buffer containing 100 ng of probe. Post-hybridization washes (1 × 2x SSC 5 min, 1 × 2x SSC 10 min) were followed by treatment with RNAse (10 μg/mL; Fisher, #BP2539) for 15 min at 37°C. Following 2 min × 5 min washes in 2x SSC, sections were placed in 0.5x SSC for 10 min, followed by a 30 min incubation in 0.5x SSC at 56°C. After 2 min × 5 min washes at RT in 0.5x SSC, sections were incubated in 2x SSC containing 3% hydrogen peroxide to inhibit endogenous peroxidases. Sections were then washed 2 × 2x SSC for 5 min before being placed into TBS (0.01 M Tris-HCL, 0.1 M NaCl, pH 7.5) for 5 min. Sections were blocked with 2% blocking buffer (Roche, #11096176001) in TBS, containing 5% goat serum for 30 min, followed by a 2 h incubation in anti-DIG-HRP (1:200; Perkin Elmer, #NEF832001EA) in blocking buffer. Sections were then washed 3x in TBS-T (TBS containing 0.05% Tween 20, pH 7.5) for 5 min, followed by a 30 min incubation with Cy3 in 1x amplification buffer (1:50; TSA Plus Cyanine 3 System, Perkin Elmer, #NEL744001KT). After 3x wash in TBS-T for 5 min each, slides were then coverslipped with Vectashield HardSet Mounting Medium (Vector Laboratories). Images were taken at 100 and 400x magnification (Olympus BX51 microscope, Olympus DP71 Digital Camera and DP manager software). Images for each gene were captured at the optimal exposure for each magnification. Images for Pri-miR132 were captured at the same exposure for both the 2 and 4 h time points. Quantification was performed on three individual animals across three slices each, spaced 80 μM apart. ImageJ (U.S. National Institutes of Health, Bethesda, MD, United States, http://imagej.nih.gov/ij/) was used to obtain the mean optical density (mean OD) of 400x images. The mean OD of the granule cell layer (GCL) or ML were compared to the region immediately below the ML (lateral posterior thalamic nucleus). The values for the stimulated and control side were then compared. The values were displayed as the fold change from the control side.

Rats were perfused with 4% PFA in 1x PBS. The NeuN IHC was performed on free-floating coronal rat sections (40 μm). Sections were blocked in donkey serum in 1x PBS for 1 h at RT. They were then incubated overnight in an anti-NeuN antibody (1:500; Millipore #MAB377). Sections were washed with PBS for 10 min, 3x before being incubated with the secondary for 2 h at RT (1:200; Invitrogen #T6390). There were 3, 10 min final washes with PBS before being mounted onto superfrost slides (Fisher) and a fluorescent mounting media containing DAPI (Vectashield).

Statistics for the image quantification, microarray and qRT-PCR analysis data were done using two-tailed t-test assuming equal variances.

One of the best methods for determining if specific mRNAs localize to dendrites is by in situ hybridization. In situ hybridization on hippocampal coronal tissue sections is especially ideal because of the unique neuroanatomical organization of the hippocampus: neuronal cell bodies are localized to discrete layers wherein the dendrites from these cell bodies project uniformly away from the soma and, therefore, spatially separate the cell body from the dendrites. These features of the hippocampus allowed for the serendipitous discovery that Arc mRNA localizes to dendrites (Link et al., 1995; Lyford et al., 1995) following neural activity of dentate gyrus neurons. Stimulating the angular bundle of the pp-HFS is one way to induce neural activity within the hippocampus, and specifically the dentate gyrus. This method is ideal because it utilizes a pattern of stimulation that produces highly reliable and robust alterations in synaptic plasticity (Bliss and Lomo, 1973; Malenka and Nicoll, 1999; Schafe et al., 2001; Rodrigues et al., 2004), and leads to robust gene expression within dentate gyrus granule cells. To illustrate this, we applied pp-HFS to urethane-anesthetized rats. One hour later, the brains were dissected and in situ hybridization was performed on coronal brain sections containing the dorsal hippocampus with radiolabeled ribo-probes for Arc mRNA and Nurr1 mRNA – two immediate early genes. As expected, both Arc mRNA and Nurr1 mRNA are robustly increased within the granule cell layer (GCL) of the dentate gyrus on the ipsilateral side of stimulation. In contrast, the contralateral side exhibits very low levels of these mRNAs. However, Arc mRNA is also detected in the molecular layer (ML) of the dentate gyrus on the ipsilateral side of stimulation, indicating that this mRNA is transported to the dendrites contained within the ML (Figure 1A). Notably, the ML consists of primarily dendrites, neuropil and traversing axons, but is virtually devoid of neuronal cell bodies. We performed immunohistochemistry for the neuronal marker NeuN, on coronal brain sections containing the dorsal hippocampus to demonstrate this. NeuN signal was very prominent in the neuronal cell bodies within the GCL of the dentate gyrus. However, there was a clear lack of NeuN stained cells within the ML (Figure 1B).

Next, we sought to identify all known transcripts within the rat dentate gyrus that are transported to the distal dendrites following synaptic activity induced by pp-HFS. Synaptic activity within the dentate gyrus of urethane-anesthetized rats induced by pp-HFS should lead to a robust increase in gene transcription of 100s of genes within GCL neurons, and we hypothesized that some of these mRNAs might be transported to the dendrites contained within the molecular layer of the dentate gyrus. These structures are so small, that accurate manual dissection would likely be impossible. Therefore, we utilized laser microdissection to dissect out the distal 2/3rds of the ML so we could ensure that we would avoid GCL contamination during the isolation of the ML (Figure 1C).

We stimulated the ipsilateral side (side of stimulation) with pp-HFS and then sacrificed the animals at 2 and 4 h post-stimulation. Cryocut coronal sections containing the anterior dorsal hippocampus were collected and subjected to Arc immunohistochemistry to verify that Arc gene expression was induced within the ipsilateral dentate gyrus (Figure 1D). We used this approach as an indirect way to verify the stimulation procedure was successful at inducing neural activity within the dentate gyrus for each animal that received pp-HFS. Next, additional cryocut sections were obtained from the dorsal hippocampus of animals that successfully received the pp-HFS. The ipsilateral and contralateral (unstimulated) distal 2/3rds of the molecular layer of the dentate gyrus was microdissected and the RNA was purified and amplified in preparation for DNA microarray analysis. We chose to analyze mRNA levels at 2 and 4 h post-stimulation compared to unstimulated controls because these are time points at which mRNAs would likely localize to dendrites following stimulation. For example, it has been previously demonstrated that Arc mRNA can be detected within the dendrites after only 30 min following stimulation and continues to accumulate within the dendrites for at least 2 h post-stimulation (Steward et al., 1998). During this period, numerous transcription factors are transcriptionally induced and promote transcription of a multitude of genes during additional waves of transcription. Two and four hours post-stimulation should be ideal time points to allow pp-HFS-induced genes to be transcribed and also allow enough time for dendrite-destined transcripts to accumulate within this compartment.

Our genome-wide screen, using Affymetrix DNA microarrays examined the expression levels of over 27,000 unique transcripts. After filtering data for changes of threefold or more, 31 transcripts were identified that were higher within the ML of the stimulated side (p < 0.05); one RNA exhibited a greater than threefold decrease in levels. Twenty-two of the 31 transcripts that were identified to be upregulated within the ML following pp-HFS, have been previously identified to be regulated via pp-HFS (Yamazaki et al., 2001; Matsuo et al., 2000; Ploski et al., 2010; Ryan et al., 2011, 2012). Eighteen of these 31 transcripts were also previously identified to be differentially regulated in neurons grown in culture following KCl-mediated induction of neural activity (Lin et al., 2008). The one transcript we identified to be decreased within the ML following pp-HFS (coiled-coil domain containing 177; Ccdc177), to our knowledge has not been previously identified to be regulated by pp-HFS, or KCl-mediated induction of neural activity. These types of studies have consistently identified a good number of genes that exhibit greater mRNA fold changes than Arc, so the fact that our screen identified Arc as the transcript with the highest increase in mRNA fold change within the ML 2 and 4 h following pp-HFS, is a good indication that the screen successfully identified mRNAs which have become enriched within the ML following pp-HFS (Figure 2). Annotated gene lists, including information regarding which genes were previously identified are provided in Supplementary File S1.

Another way to examine the microarray data is to view each gene’s signal intensity on the microarray comparing the control, 2 h, and 4 h samples (Figure 2D). Most of these genes exhibit fairly low signals on the microarray, which could indicate their overall levels are low in the ML. In contrast, Arc mRNA exhibits the highest signal intensity with only a few other mRNAs exhibiting similar levels.

We next performed qRT-PCR for a subset of the identified candidate genes that either exhibited the greatest fold change or exhibited a strong signal on the microarray (Figure 3). For these experiments, we utilized the aRNA that was generated for the microarrays, converting it to cDNA and then performed qRT-PCR with gene specific primers. Each gene we examined exhibited significant increases in RNA levels compared to controls (with the exception of pri-miR132 at 2 h) (n = 3 per group; p < 0.05). Statistical data for these qRT-PCR experiments are provided in Supplementary File S3.

The second most robustly upregulated transcript at 4 h post-stimulation was a partially characterized transcript referred to as RM2 (GenBank: AB032083.1), which had reportedly been previously identified to be induced within the hippocampus following LTP inducing stimulation (Matsuo et al., 2000). There was no evidence that RM2 contained a reading frame to code for a protein. We reasoned that this transcript could be part of a larger, as of yet, unidentified transcript. Therefore, we examined the rat genome upstream from the RM2 genomic locus and we determined that a CREB dependent non-protein coding transcript that codes for the miRNAs, miRNA-212 and miRNA-132 (Vo et al., 2005) was directly adjacent to the RM2 locus. We hypothesized that the RM2 sequence might be part of a contiguous RNA that contains miRNA 212 and 132. To test this hypothesis, we subjected cDNA generated from RNA obtained from the distal molecular layer, 4 h post HFS (similar material that was used for the microarray analysis), to PCR with PCR primers specific for the RM2 microarray probe and the miRNA-212 or miRNA-132 loci. The PCR did not amplify a DNA product using the miRNA212 primer and RM2 primer, but it did amplify two PCR products that were ∼1.6 kb and ∼660 bp utilizing the miRNA132 and RM2 primers. These PCR products were cloned and subjected to DNA sequencing for verification. We determined that indeed RM2 was contiguous with an RNA transcript that contained miRNA132 (Figure 4). The ∼660 bp transcript was shorter than the ∼1.6 kb transcript because it was lacking a region intervening between miR132 and RM2, most likely due to the presence of an intron that was spliced out. These data are consistent with findings published while this work was ongoing (Takebayashi et al., 2014). Collectively, these findings are intriguing because miRNA132 is believed to localize to dendrites (Edbauer et al., 2010; Bicker et al., 2014), but it is currently unknown how it gets there (Tai and Schuman, 2006). Conventional wisdom dictates miRNAs are expressed within the nucleus as part of larger RNA polymerase II transcripts (pri-miRNAs), and these are cleaved within the nucleus by Drosha and DGCR8 to create pre-miRNAs, which are then exported to the cytoplasm for further processing by DICER to convert pre-miRNAs to mature miRNAs (Corbin et al., 2009; Bicker et al., 2014). However, our data indicate there is a large upregulation of pri-miRNA132/RM2 within the molecular layer (dendrites) following synaptic activity, indicating that pri-miR132 may leave the nucleus unprocessed. This possibility makes it plausible that miRNA132 is transported to the dendrites as a pri-miR132. Recent reports lend credibility to this model. For example, pri-miRNAs, Drosha and DGCR8 have recently been reported to be present within the post-synaptic density (PSD) (Lugli et al., 2012). These data indicate that miRNA132, might be transported to the dendrites following synaptic activity as a longer unprocessed transcript which is then likely processed to a mature miRNA when it reaches its destination. We will refer to RM2 as pri-miR132.

We stimulated another group of animals with pp-HFS and 2 and 4 h later, the brains were dissected and coronal sections containing the dorsal hippocampus were taken. We performed secondary confirmation on a subset of candidate genes utilizing FISH using gene specific riboprobes (Figure 5). We chose to examine mRNA expression for the selected genes for the 4 h time point following pp-HFS, since this was the timepoint these genes exhibited the highest expression within the ML, according to the microarray data. For pri-miR132, we also examined expression at the 2 h time point. As expected, FISH performed for Arc yielded a robust and very convincing signal for Arc within the GCL and ML on the stimulated side of the brain. In contrast, the unstimulated, control side exhibited very little signal within these regions. We also examined the expression pattern for Egr3, Egr4, H1a, pri-miR132, Sox11, and Ptgs2 (n = 3). In all of these cases, these genes exhibited a significantly elevated expression signal within the GCL on the stimulated side of the brain as compared to the unstimulated hemisphere. Sox11, a known neurogenesis marker (Bergsland et al., 2006), is seen expressed in small subsets of cells within the subgranular zone within the GCL of the unstimulated side, but the stimulated side appears to express Sox11 in all of the GCL cells similar to the other mRNAs we examined. In contrast, there was generally little or no signal detected within the ML for these genes. Egr3, Egr4, and H1a exhibited a modest, but statistically significant signal within the ML on the stimulated side. These three mRNAs also appeared to be expressed within the hilus on the stimulated side. None of the genes appeared to be expressed by specific cells localized within the ML (i.e., glia). If the mRNAs for pri-miR132, Sox11, and Ptgs2 do indeed localize to dendrites within the ML, they likely do so at much lower levels than Arc mRNA, considering that our ability to detect them within the ML using FISH is limited. We also plotted the intensity of the FISH signal for these RNAs over the GCL and ML (Figure 6). Arc mRNA exhibits a robust GCL and ML signal. The other RNAs examined exhibit a much more modest ML signal as expected; however, these data indicate that Egr3 appears to extend further into the proximal dendrites. Statistical data for these FISH experiments are provided in Supplementary File S3.

In our final set of experiments, we sought to determine if utilizing a more sensitive approach could detect pp-HFS-induced mRNAs within the ML. We stimulated another group of animals with pp-HFS and then 2 and 4 h later, the brains were dissected and coronal sections containing the dorsal hippocampus were prepared for laser microdissection. This time we dissected tissue from the GCL from both the stimulated and unstimulated sides of the brain. We also dissected tissue from the distal most 1/3rd of the ML from both the stimulated and unstimulated sides of the brain. The RNA was isolated from this tissue and converted to cDNA and subsequently analyzed via qRT-PCR with gene specific primers (Figure 7). At 2 h post pp-HFS, Arc mRNA was detected in the GCL and ML on the stimulated sides compared to the unstimulated sides, as expected (p < 0.05; n = 4 for 2 h, n = 5 for control). H1a, pri-miR132, Sox11, BDNF, Egr2, and Egr4 exhibited an increase in the GCL on the stimulated side compared to the unstimulated side (p < 0.05; n = 4 for 2 h, n = 5 for control). However, for this time point, none of these targets exhibited an increase in the ML on the stimulated side compared to the unstimulated side (p > 0.05; n = 4 for 2 h, n = 5 for control), except for pri-miR132 and Egr4 (p < 0.05; n = 4 for 2 h, n = 5 for control). Notably, Egr4 exhibited a large increase in the ML, with a C_t difference of ∼4 or ∼16 fold. At 4 h post pp-HFS, Arc mRNA was detected in the GCL and ML on the stimulated sides compared to the unstimulated sides, as expected (p < 0.05; n = 4 for 4 h, n = 5 for control). H1a, pri-miR132, Sox11, BDNF, Egr3, and Egr4 exhibited an increase in the GCL on the stimulated side compared to the unstimulated side for this time point (p < 0.05; n = 4 for 4 h, n = 5 for control). The qRT-PCR also detected an increase for H1a, pri-miR132, Sox11, Egr3, and Egr4 within the ML on the stimulated side compared the unstimulated side for the 4 h time point (p < 0.05; n = 4 for 4 h, n = 5 for control). Interestingly, there wasn’t an appreciable increase in BDNF in the ML on the stimulated side compared to the non-stimulated side of the brain. Statistical data for these qRT-PCR experiments are provided in Supplementary File S3.

The qRT-PCR data allowed us to examine the relative levels of these mRNAs within the dentate gyrus, which could provide some insight into how these differing mRNAs compare to the levels of Arc mRNA. To do this, we first normalized the levels of each RNA to the levels of GAPDH mRNA. We then compared the control GCL sample Arc mRNA levels to the other control GCL mRNA samples examined (i.e., Egr2, Egr3, etc.). We found that Arc mRNA levels were modestly higher than the other mRNAs examined except for Egr3 (Figure 8A); however, when we compared Arc GCL mRNA levels at the 2 and 4 h timepoints to the other GCL mRNAs examined at these time points, we found that Arc levels were much higher than the other mRNAs for these time points. We then compared the control sample’s Arc mRNA levels within the ML to the other control ML mRNAs. We found that Arc ML mRNA levels were higher than the other ML mRNAs examined (Figure 8B). When we compared the 2 and 4 h timepoint Arc mRNA levels within the ML to the other mRNAs examined for the 2 and 4 h time points, we found that Arc ML levels were significantly higher than the other mRNAs. We report the relative fold differences of Arc mRNA compared to the other mRNAs examined in the GCL and ML (Figures 8C,D). Arc mRNA generally is more abundant than the other mRNAs examined. Within the ML compartment, Arc levels can exceed the levels of some of these other mRNAs 100–1000-fold. In our last analysis, we compared the GCL versus ML levels of each mRNA (Figures 9A,B). We found that Arc mRNA levels were consistently higher within the ML compared to the GCL across all samples (control, 2 h, and 4 h). In contrast, virtually all the other mRNAs examined exhibited much higher GCL levels compared to ML levels. An important exception was Egr4 at the 4 h time point. Collectively, these mRNA level findings corroborate the DNA microarray findings indicating that Arc mRNA levels are much higher within the ML compared to most of the other mRNAs we examined. Additionally, these data support the notion that while some of these mRNAs do increase their levels within the ML following pp-HFS, their relative levels compared to Arc mRNA are exceedingly low and this is likely the main reason why FISH analysis was unable to convincingly detect these candidate mRNAs within the ML.